-
Varicam 24p: clip TC vs. winowburn TC and cadence discepancies
I’m cutting a show where the majority of the footage is shot on Varicam and am experiencing some problems (or at least some uncertainties). This is the first show I’ve cut using the Varicam so am not entirely familiar with the intricacies of the workflow. Here’s what’s going on:
* Our production crew shot two-camera interviews with Varicams at the 24p setting and using “time of day” timecode for sync between A and B cameras.
* I captured these tapes into Final Cut Pro 6 using HD-SDI output from AJ-HD1200A and into SDI input on Kona-2 card.
* Used 9-pin machine control with “AJA Kona2: 720p 59.94 Varicam Panasonic VTR” device control preset.
* Captured footage using the “AJA Kona2: 720p 23.98 DVCPro HD Varicam” capture preset.
* I put a TC window burn on video coming out of deck.In the end, we will online the show and master to 24p HDCAM. In addition to the Varicam footage, we will be mixing in footage shot with HVX200 on P2 cards and some archival.
Here’s what concerns me. In the clips I’ve captured, there are discrepancies between the clips’ timecode and the window burn timecode:
* The “cadence” doesn’t skip the same numbers. For instance, frames in the window burn might go 00, 01, 02, 03, 05, 06, 07, 08, 10.. etc. while theclip’s timecode will go 00, 02, 03, 04, 05, 07, 08, 09, 10… etc.
* While most of these clips’ timecode cadence will “line up” with the window burn cadence after every 4 frames, it doesn’t always happen on the same frame — in the above example, it lines up on the “00” frame, but sometimes it will line up on “18” or “15” or whatever.
* A few clips don’t “line up” at all — the clip timecode and the window burn remain at least 1 frame out of sync, even where the other clips “line up” (the window burn reads “00” while the TC reads “01”)
I don’t want to get into the online and be worrying about the audio drifting 1 or 2 frames. My initial impulse was to recapture at 29.97 and sort it out with Cinema Tools (I’m used to working with film). I used the Media manager to “create offline” and recapture at 29.97, but found that the cliptimecode and the window burn were now wildly off, 20 frames up to a few minutes in some cases!
I’ve heard that the Varicam footage needs to be captured starting on the “00” frame, but it seems to me that shouldn’t make a difference if the Varicam “flags” the extraneous frames in 24p.
Can anyone offer any insight? I’m looking for peace of mind before I commit to recapturing this stuff that I won’t be dealing with sync issues in the online.
-
10 Replies
-
Multicam footage must be genlocked with proper TriLevel Sync in order for the time code to slave to be completely accurate. I suspect this is your problem. Check with the camera crew to see if this was done!
JS
-
[Wheelhouse] ” I put a TC window burn on video coming out of deck.”
then that video is NOT 24p it is 60p- the deck cannot output 23.98 without putting it back into a 60p video stream.
“Here’s what concerns me. In the clips I’ve captured, there are discrepancies between the clips’ timecode and the window burn timecode:
* The “cadence” doesn’t skip the same numbers. For instance, frames in the window burn might go 00, 01, 02, 03, 05, 06, 07, 08, 10.. etc. while theclip’s timecode will go 00, 02, 03, 04, 05, 07, 08, 09, 10… etc. “did you calibrate the capture offset in FCP? This is exactly what it is used for – aligning up TC so that everything matches.
finally- most decks do not see 60 frames only 60 fields and they use TC counting for 29.97 make sure that the fields indicators are in the TC burn on so that you actually know when you are on the first frame/ field 00, 01, 02 or you are looking at the 2nd frame/ field 00*, 01*,02*
and yes (like JS said) the cameras need to be genlocked to make sure they are all on the same time.
gary adcock
Studio37
HD & Film Consultation
Post and Production Workflows -
Thanks to all for taking the time to respond. I’m still confused, so I’m going to be more focussed in my question.
First, I’m not all that concerned about synch between A & B cameras. I know they didn’t synch the cameras properly, but I’ll resolve that later.
Second, I’m aware that the video coming out of the deck is 60p. It’s my understanding the “flagged” frames are removed during capture. The video I’ve ingested is certainly 23.98 and there are no “repeated” frames (like you might see on an imporperly done reverse telecine from film footage).
Third, the offset should be okay — for most of the footage, the clip TC window and the Window Burn TC agree every four frames. The clip TC and windowburn TC are not “out of synch,” they’re “out of step.”
What’s happeniing is this.
Frame 1
ClipTC WBTC
00 00Frame 2
ClipTC WBTC
02 01Frame 3
ClipTC WBTC
03 02Frame 4
ClipTC WBTC
04 03Frame 5
ClipTC WBTC
05 05etc.
Here’s the heart of my question: when prep I the EDL for the online, what do I believe, the windowburn TC or the clip TC? What’s causing these two TC values to be out of step?
Unfortunately, I don’t have the deck right now to expirement with whether I am seeing an odd or even “field.” But I’d like to understand what I’m seeing before renting it again and redig-ing.
Thanks again in advance for your help.
-
[Wheelhouse] “Third, the offset should be okay — for most of the footage, the clip TC window and the Window Burn TC agree every four frames. The clip TC and windowburn TC are not “out of synch,” they’re “out of step.””
did you turn on the “field” indicator in the deck?
Since Panasonic uses the industry standard for TC at 29.97 you are looking at only 1/2 of the 60 fps when viewing on a deck without the field indicator on. Since most NLE’s are not capable of editing at the field level – the number is rounded to the next frame.
as long as the in and out point are correct – your edl will be a good as one can be.
gary adcock
Studio37
HD & Film Consultation
Post and Production Workflows -
Do you know if the crew used an external TC generator or used Cam A as the generator? If you do not use one of the cameras as a generator I believe they will not be in the same pulldown cadence.
-
Thanks. Final Cut rounding to the next field seems like a completely plausible explaination for what I’m seeing. I’ll check it out for sure when we rent the deck again in a couple weeks.
-
The crew didn’t use any kind of synch other than “time of day.” But again, I’m much less worried about synch between the two cameras at this point than I am any given clip’s timecode being accurate for the EDL.
-
Yes, just as I suspected, no trilevel sync connection! What this really means is that your EDL must contain the discreet TC from the picture that accompanies it. No problem, unless of course you’re cutting to a common sound track with the code of one and not the other camera. Of course at worst you’ll be a frame off on the second “b-roll” camera.
My advice is to proceed normally and carefully observe the result of the online and adjust the one frame discrepancy if it’s noticeable, which again, is unlikely.
JS
-
Okay, let me get this straight. To capture 24p footage shot on a varicam at 59.94 using a panasonic AJ-HD1200A (also running at 59.94) and a Kona-2 card (which only captures the 24 “flagged frames” in every second of footage), the Kona and the 1200A need to be synched with tri-level sync? I was under the understanding that since the tape itself is running at 59.94 during capture, one could sync with standard blackburst. Is that the source of my issues?
-
No. No sync necessary for digitizing. I’m talking about the cameras needing genlock (trilevel hi def sync) in order for the code to be locked; in other words for the cadence to be locked.
JS
Reply to this Discussion! Login or Sign Up